From e7905a9395c1a002f50bab29e16a729e14d4ed6f Mon Sep 17 00:00:00 2001 From: Eric Liang Date: Thu, 30 Jul 2015 16:15:43 -0700 Subject: [SPARK-9463] [ML] Expose model coefficients with names in SparkR RFormula Preview: ``` > summary(m) features coefficients 1 (Intercept) 1.6765001 2 Sepal_Length 0.3498801 3 Species.versicolor -0.9833885 4 Species.virginica -1.0075104 ``` Design doc from umbrella task: https://docs.google.com/document/d/10NZNSEurN2EdWM31uFYsgayIPfCFHiuIu3pCWrUmP_c/edit cc mengxr Author: Eric Liang Closes #7771 from ericl/summary and squashes the following commits: ccd54c3 [Eric Liang] second pass a5ca93b [Eric Liang] comments 2772111 [Eric Liang] clean up 70483ef [Eric Liang] fix test 7c247d4 [Eric Liang] Merge branch 'master' into summary 3c55024 [Eric Liang] working 8c539aa [Eric Liang] first pass --- R/pkg/inst/tests/test_mllib.R | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'R/pkg/inst/tests/test_mllib.R') diff --git a/R/pkg/inst/tests/test_mllib.R b/R/pkg/inst/tests/test_mllib.R index 3bef693247..f272de78ad 100644 --- a/R/pkg/inst/tests/test_mllib.R +++ b/R/pkg/inst/tests/test_mllib.R @@ -48,3 +48,14 @@ test_that("dot minus and intercept vs native glm", { rVals <- predict(glm(Sepal.Width ~ . - Species + 0, data = iris), iris) expect_true(all(abs(rVals - vals) < 1e-6), rVals - vals) }) + +test_that("summary coefficients match with native glm", { + training <- createDataFrame(sqlContext, iris) + stats <- summary(glm(Sepal_Width ~ Sepal_Length + Species, data = training)) + coefs <- as.vector(stats$coefficients) + rCoefs <- as.vector(coef(glm(Sepal.Width ~ Sepal.Length + Species, data = iris))) + expect_true(all(abs(rCoefs - coefs) < 1e-6)) + expect_true(all( + as.character(stats$features) == + c("(Intercept)", "Sepal_Length", "Species__versicolor", "Species__virginica"))) +}) -- cgit v1.2.3